Virgo Logbook

Virgo Runs (O3)

magazzu - 23:04 Tuesday 03 March 2020 (48607)

Operator Report - Afternoon shift

ITF found in relocking phase, after a series of failed attempts caused by the strong wind we were able to reach LOW_NOISE_3_SQZ at 15:11 UTC. Thanks to the configuration prepared by Allocca the ITF was already locked in EQ Mode; in agreement with the Run Coordinator the ITF was set in Science Mode at 15:15 UTC while in this configuration.
During this lock the SQZ disengaged two times ( 15:56 UTC, 16:05 UTC) possibly due to the bad weather, in both case it was possible to go back in Science Mode less then a minute later.
ITF unlocked at 16:18 UTC due to a Fast Unlock ( see attached plot #1 ). After some attempts we were back in LOW_NOISE_3_SQZ and EQ Mode at 16:49 UTC.
After that the wind activity was decreased we switched to Adjusting Mode from 17:50 UTC to 17:52 UTC to disable the EQ Mode, no issue encountered in this phase even with the custom configuration.
ITF left in Science Mode.

Sub-system reports

DAQ
From around 19:15 UTC the process DDS1 started to be shown as red on the VPM. It then return for few minutes back in its normal state to switch back to red soon after. I tried to stop and restart it manually but 21:12 UTC but this action did not solve the problem. Expert informed of this behavior.

SQZ
From 14:15 UTC to 14:25 UTC the DAQ Oncall ( Pacaud ) solved the Timing issue related to the Squeezer. After that I restored the standard configuration of SQZ_MAIN.ini as indicated by the pending action.

Pending actions

ISC
(03-03-2020 22:00 - ) At the next unlock reload ITF_LOCK to lock without EQ Mode.

Images attached to this report

48607_1583273040_Screenshot from 2020-03-03 23-02-18.png

Comments to this report:

carbognani - 10:27 Wednesday 04 March 2020 (48610)

>From around 19:15 UTC the process DDS1 started to be shown as red on the VPM. It then return for few minutes back in its normal state to switch back to red soon after. I tried to stop and restart it manually but 21:12 UTC but this action did not solve the problem. Expert informed of this behavior.

This morning, in an attempt to solve the problem:

the corresponding tango server std2011/DDS1 has been stopped/restarted
the ethernet bridge used for the communication (ethcal.virgo.infn.it) has been reeboted

This did not solve the problem and large delay in reading from the device remain, as shown by monitoring the server from tango side. (See Fig. 1).

An investigation on the actual hardware in the INJ EERoom is needed and will be performed at the next occasion.

Images attached to this comment

rolland - 10:36 Wednesday 04 March 2020 (48612)

Confirmation that EQ mode in compatible with Science mode (from the point of view of h(t) reconstruction).

The plot shows some monitoring channels during the EQ mode of yesterday afternoon, and at and after the transition EQ mode to normal mode at 17h50 UTC.

Compared to tests done in January and presented in VIR-0097A-20, the artifact of having h(t) = 0 at the transition when the EQ mode is stopped has been removed (thanks to permanent calibration lines added to NI and WI marionettes).
The monitoring of hrec bias at frequencies of different hardware injections are also shown (modulus and phase). They are stable in time, independent of the mode as it should be. It confirms that hrec is behaving normally while ITF is in EQ mode (thanks to subtracting the NI and WI marionette Z_CORR signals in Hrec).

Images attached to this comment

carbognani - 10:27 Thursday 05 March 2020 (48623)

Actually the DDS box on the INJ EERoom was switched off since not used anymore. In the current situation, the 22Mhz is generated with the second LNFS and because of that the DDS was unconnected but still on.

I have stopped the corresponding servers and removed the DDS1 instances on the VPM and on Tango and now the VPM Injection flag is back green on the DMS.